对比10种模型预测新英格兰电力需求，发现疫情数据会因行为变化过时。
Short-Term Electricity Demand Forecasting for New England: A Comprehensive Machine Learning Benchmark with Weather, Calendar, and COVID-19 Indicators
- 用十类机器学习模型融合天气、日历和疫情数据做电力预测
- CatBoost表现最佳，疫情特征移除后测试误差降1.7%
- 疫情数据在2022年8月后已失效，模型仍过度依赖旧模式
准确的短期电力需求预测对电网运行、能源市场规划和基础设施优化至关重要。本文在2020年2月至2023年3月期间，针对新英格兰ISO区域对十种机器学习模型进行基准测试。模型涵盖四类：表格型梯度提升树（随机森林、LightGBM、CatBoost、XGBoost）、独立神经架构（LSTM、Transformer编码器）以及混合式Transformer+树模型（Hybrid XGBoost、Hybrid LightGBM、Hybrid CatBoost、Hybrid RF）。所有模型均使用六个城市的气象数据、日历与节假日效应、自回归需求滞后项及新冠流行病学变量。超参数优化采用Optuna（300次试验，多变量TPE，seed=42），按70/15/15时间顺序划分数据集且无信息泄漏。CatBoost表现最优：测试集RMSE为8316 MWh，MAPE为1.87%，R-squared为0.917；其次为XGBoost（9066 MWh，R² 0.901）、Hybrid CatBoost（9068 MWh，R² 0.901）和Hybrid XGBoost（9208 MWh，R² 0.898）。独立神经网络性能显著较差（Transformer：21294 MWh；LSTM：22808 MWh），证实Transformer更适合作为特征提取器而非端到端预测器。对CatBoost的消融实验表明，需求滞后项是主导因素：移除后RMSE从8316升至11310 MWh（+36%），而仅保留天气与日历特征即可达R² 0.864。移除新冠特征使测试集RMSE降低1.7%，训练集RMSE下降17.3%，表明存在时间有效性衰减。SHAP分析验证此现象：8个新冠特征中，有3个在疫情活跃期训练集排名更高，但在后急性期测试集排名上升，说明模型仍过度应用已失效的疫情模式。
原文摘要 · Abstract (English)
Accurate short-term electricity demand forecasting is critical for reliable power system operation, energy market planning, and infrastructure optimization. This paper benchmarks ten machine learning models for daily electricity demand forecasting across the New England ISO (February 2020 - March 2023). The models span four families: tabular gradient-boosted trees (Random Forest, LightGBM, CatBoost, XGBoost), standalone neural architectures (LSTM, Transformer encoder), and hybrid Transformer+tree variants (Hybrid XGBoost, Hybrid LightGBM, Hybrid CatBoost, Hybrid RF). All models use meteorological data from six cities, calendar and holiday effects, autoregressive demand lags, and COVID-19 epidemiological variables. Hyperparameter optimization uses Optuna (300 trials, multivariate TPE, seed=42) under a leakage-free 70/15/15 chronological split. CatBoost achieves the best test performance: RMSE 8316 MWh, MAPE 1.87%, R-squared 0.917, followed by XGBoost (9066 MWh, R-squared 0.901), Hybrid CatBoost (9068 MWh, R-squared 0.901), and Hybrid XGBoost (9208 MWh, R-squared 0.898). Standalone neural architectures perform substantially worse (Transformer: 21294 MWh; LSTM: 22808 MWh), confirming the Transformer's role as a feature extractor rather than an end-to-end forecaster. An ablation on CatBoost shows that demand lags are the dominant predictor: removal degrades RMSE from 8316 to 11310 MWh (+36%), while weather and calendar features alone achieve an R-squared of 0.864. Removing COVID-19 features improves test RMSE by 1.7% while reducing training RMSE by 17.3%, a signature of temporal validity decay. SHAP analysis confirms this: 3 of 8 COVID features rank higher on the post-acute test set than during pandemic-active training, indicating the model over-applies stale pandemic patterns after behavioral adaptation was complete by August 2022.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。